Driving T cells to human atherosclerotic plaques: CCL3/CCR5 and CX3CL1/CX3CR1 migration axes

Abstract

T-cell accumulation in atherosclerotic plaques contributes to plaque destabilization. We found that several chemokine receptors are differentially expressed on peripheral blood compared to plaque-resident T cells and corresponding ligands are upregulated in plaques. These data indicate that T-cell migration into human atherosclerotic plaques may predominantly occur via CCR5-CCL3 and CX3CR1-CX3CL1 interactions. The linkage of atherosclerosis to immune activation can be traced back as far as to Rudolf Virchow [1]. Cytokine release and alteration of surface marker expression on immune cells reflect this activation. Disorders characterized by a proinflammatory environment have been associated with an increased risk of atherosclerosis and many works implicate adaptive immunity in this risk [2, 3]. Advanced atherosclerotic plaques are enriched for activated T cells, thus pointing to their potential role in plaque progression and destabilization [4]. Nonetheless, the mechanisms of T-cell accumulation in plaques remain to be elucidated. To gain insight into these mechanisms, we analyzed the surface phenotypes of T cells isolated from human peripheral blood and autologous atherosclerotic plaques. Generally, immune cell phenotypes found in the plaques in our study are in agreement with those recently reported in mass cytometry (CyTOF) and single-cell RNA sequencing studies [5, 6]. Specifically, among plaque CD4+ and CD8+ T cells, the predominant population had a CD45RO+CD27- effector memory phenotype (Fig. 1A and C), which is associated with viral infection and/or other chronic inflammatory states that trigger T-cell activation. Accordingly, proportions of plaque-resident T cells expressing the early activation marker CD69 were increased (Fig. 1B and D). CD69 blocks tissue egress of T cells to blood according to a sphingosine-1-phosphate gradient. Thus, elevated surface CD69 expression may provide a mechanism for T-cell retention within the atherosclerotic plaque by blocking their exit. But what drives T cells to the plaques? Unlike most previous works on human atherosclerotic plaques that focused on T-cell heterogeneity or activation/exhaustion status, we addressed the potential migratory mechanisms that may drive T cells from the bloodstream into the vessel wall. Thus, we analyzed the chemokine receptors that are differentially expressed on T cells in peripheral blood and atherosclerotic plaques. CCR2 (where CCR is CC (cysteine-cysteine) chemokine receptor) was not found at substantial levels on CD4+ and CD8+ T cells, while T cells expressing CCR4, CCR5, CCR7, and CX3CR1 (where CX3CR is CC (cysteine-XXX-cysteine) chemokine receptor) were differentially represented in plaques and blood (Fig. 1B and D). The fraction of CCR5+ cells in plaques was significantly higher, indicating the possible involvement of this surface molecule in T-cell migration into the plaques. In contrast, fractions of CCR4+, CX3CR1+, and CCR7+ cells were lower in plaques than in blood (Supporting Information Tables S3 and S4). Low expression of CCR7 on plaque T cells was expected since this receptor is not expressed by tissue-resident cells. However, CCR4 and CX3CR1 decrease could result from their rapid internalization upon binding to corresponding ligands. Indeed, CCL2 (where CCL is CC (cysteine-cysteine) chemokine ligand), CCL3, and CX3CL1 mRNAs (where CX3CL is CC (cysteine-XXX-cysteine) chemokine ligand) that encode known ligands to CCR4, CCR5, and CX3CR1, respectively, were increased in plaques compared to their levels in blood leucocytes (Fig. 2). Next, we analyzed chemokine receptor turnover upon binding to their plate-bound ligands (Supporting Information Fig. S4). Exposure to CCL2 did not result in CCR4 removal from the T-cell surface, whereas CX3CL1 exposure did lead to CX3CR1 removal. Importantly, in contrast to CCL4 and CCL5 exposure that diminished surface CCR5 levels, CCL3 exposure did not alter CCR5 surface levels. In contrast to above-mentioned experiments, data on surface chemokine receptor clearance driven by soluble chemokines (e.g., [7]), confirmed in our study (data not shown), indicate that soluble CCL3 can drive CCR5 clearance. However, we suppose that plate-bound experiments reflect the in vivo situation more faithfully since in plaques chemokine receptors on T cells seem to interact with ECM-bound rather than with soluble chemokines (discussed in [8, 9]). Our in vitro surface chemokine receptor clearance experiments indicated that the low expression of CCR4 on plaque-resident T cells is not a consequence of its removal upon ligand exposure. In contrast, sustained surface expression of CCR5 and loss of CX3CR1 observed after exposure to CCL3 and CX3CL1, respectively, are consistent with the observed phenotypes of plaque-resident T cells. This suggests that the CCL3-CCR5 and CX3CL1- CX3CR1 axes are involved in the homing of circulating T cells to atherosclerotic plaques. These data support our recent report on the importance of the CX3CL1-CX3CR1 axis for CD4+ T-cell migration into plaques [10]. We found that inhibition of CX3CL1-CX3CR1 interactions reduced cell migration toward activated endothelial cells in vitro in some but not all donors, suggesting that a second signal could also govern CD4+ T-cell traffic to endothelium. Our current findings suggest that the CCL3-CCR5 axis might provide a second migratory signal that could differentially drive T-cell subpopulations to infiltrate atherosclerotic plaques. Still, other CCR5 ligands, CCL4 and CCL5, might be involved in T-cell recruitment to atherosclerotic plaques since their expression was detected in all plaques. However, we suspect that their contribution to the total T-cell recruitment is likely to be less important than that of CCL3. Thus, among CCR5 ligands CCL3 may be a key, but not the only, driver of T-cell migration. Taken together, CCR4, CCR5, and CX3CR1 chemokine receptors were differentially expressed on human T cells from atherosclerotic plaques and peripheral blood. We showed that plaques were enriched for mRNAs encoding CCL2, CCL3, and CX3CL1, the ligands to these receptors. Our experiments on receptor turnover confirmed that the CCR4-CCR5+CX3CR1- phenotype of plaque-infiltrating T-cells may originate from the surface receptor clearance in case of CX3CR1 but not CCR4 and that the accumulation of CCR5+ T-cells in plaque could be related to heightened plaque expression of CCL3 that does not efficiently promote CCR5 removal. Thus, T-cell migration from peripheral blood into atherosclerotic plaques in humans may occur predominantly through the CCL3-CCR5 and CX3CL1-CX3CR1 interaction axes. Pharmacological disruption of these interactions may represent a useful strategy to prevent atherosclerotic plaque progression. This study was funded by Russian Science Foundation #20-75-10085 (to A.K. and D.P.); Intramural Program of National Institute of Child Health and Human Development (to L.M.); and Richard J. Fasenmeyer Foundation and the National Institutes of Health AI 068636 (to M.L.F. and M.M.L.). The authors declare no commercial or financial conflict of interest. The peer review history for this article is available at https://publons.com/publon/10.1002/eji.202049004. Data are available in article Supporting Information Materials. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts